Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    structural or mathematical representations and notation in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    mathematical or latex constructs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    PerformLayout
    -0.77
     zwiſchen
    -0.72
    bootstrapcdn
    -0.70
     dieſem
    -0.70
    transQ
    -0.69
     dieſe
    -0.69
     Савезне
    -0.69
     dieſer
    -0.68
    ViewFeatures
    -0.68
     Geiſt
    -0.68
    POSITIVE LOGITS
     gum
    0.33
     
    0.31
     (
    0.29
     bottom
    0.26
     Cha
    0.26
     neither
    0.26
     f
    0.26
     i
    0.26
    ka
    0.25
     whose
    0.25
    Act Density 0.108%

    No Known Activations